Launching the Game of Trades Model Portfolio: A Collection of Our Best Bets

Here’s the video version of the article:

Summary

• Today we’re launching the Game of Trades Model Portfolio, the product of our strategic and tactical research. Our investment process is comprised of those three components.
• The portfolio represents a collection of our best bets. It maximizes the risk-reward of our highest-conviction ideas derived from our research.
• Our biggest position is on equities, with consumer discretionary, tech and transportation stocks making up 50% of the portfolio. The bet represents peaking inflation and resilient earnings.
• Bitcoin comprises roughly 5% of the portfolio. We believe the cryptocurrency will benefit from rising stocks and a weakening dollar; a bet that the Fed will ease off.
• About 30% of the portfolio is comprised of Treasury bonds. They’re an attractively-priced hedge to our big equity exposure. They can protect us if the economy weakens more than expected.
• We have a relatively-small short position on oil, believing there’s still more downside. The big decline in prices of late, however, has cut some of that runway.

How the Model Portfolio Fits Within Our Investment Process

Over the last several months we’ve revamped our efforts to boost our research capabilities. That led us to establish a two-fold investment process that complement each other: strategic and tactical. Our Model Portfolio, the expression of our research efforts, adds a third component as shown in the diagram below.

The strategic component formulates our investment thesis, that has an investment horizon of six-to-12 months. The approach is top-down, contextualizing the macroeconomic backdrop with historical analysis. To do that we leverage our extensive analytical capabilities. We seek to understand the key dynamics operating across markets and exploit big dislocations.

The tactical component leverages our expertise in technical analysis to highlight attractive setups and structures that best express the views derived from our strategic research. We look for opportunities that maximize the risk-reward.

The model portfolio synthetizes our strategic and tactical research into a simple and transparent vehicle that expresses our bets. The objective of the portfolio is to maximize the risk-reward across the various opportunities derived from our research. The chart below depicts the portfolio’s asset allocation as of today.

In the sections that follow we’ll delve into the portfolio construction process and the underlying bets.

Portfolio Construction: Maximizing the Risk-Reward

Our Model Portfolio seeks to maximize the risk-reward ratio of the collection of ideas derived from our research. To do that we estimate expected returns (i.e., reward) and expected volatilities (i.e., risk) for each asset class, using those estimates to compute risk-reward ratios for each of them.

The expected returns (i.e., reward) are derived from our strategic research and are based on six-to-12 month investment horizons. The process for deriving expected returns varies by asset class. Some may be derived from a model that weighs in important dynamics dominating the return structure of a particular asset class. In the case of Bitcoin, for example, that involves forecasts for the S&P 500 and the dollar. In the case of the S&P 500, the expected return is a function of our forecasts for earnings growth and P/E valuation multiples.

Importantly, the expected returns are vetted by those derived from price targets sourced from our tactical research. We want to see consistency between the strategic and tactical targets, as that improves the accuracy of the forecasts. The expected volatilities (i.e., risk) comprise an average between historical volatilities, and those implied from the options market.[1]

We then input the expected returns and volatilities through a portfolio optimizer. The optimizer is meant to maximize the risk-reward of the portfolio, or in finance parlance, the Sharpe ratio (i.e., expected return less the risk-free rate, all of it divided by the volatility).[2]

Lastly, we apply constraints to the portfolio as a measure of risk management. For example, we may limit the allocation of any particular position, particularly to those with inherently high volatility, regardless of how attractive we find the opportunity. That would help prevent any single position from unduly impacting the overall performance of the portfolio.

Asset Allocation: Equities Represent Our Biggest Bet

The model portfolio is heavily-weighted toward equities, that we see as the best expression to capitalize on the incoming disinflation regime and resilient earnings growth. Our target for the S&P 500 over the next 6-12 months is $5,200. We’ll use that index as our performance benchmark.

In this section we’ll summarize the investment thesis for the eight asset classes featured in the model portfolio. The thesis for each is derived from our research.

Consumer Discretionary Stocks (Long, XLY)

In prior research we found similarities between today’s consumer and that in the years following WWII, when there was also a lot of pent-up demand and excess savings. Today the latter is about $2.5 trillion, about 16% of expected consumer spending in 2022 and 2023, a big number by the standards of history as shown below.

Moreover, spending is being underpinned by a strong labor market. We believe the decline in real goods spending so far this year has been driven by rebalancing toward services, not a sign of underlying weakness. All of this is supportive of a constructive picture for earnings.

In prior work we found consumer discretionary stocks won big in disinflation regimes. That makes sense because they’re typically sensitive to interest rates, and hence an easing Fed can be a tailwind. Consumer discretionary stocks have outperformed the S&P 500 by around 10 percentage points since we published our work on the sector.

Information Technology Stocks (Long, XLK)

The aggressive rise in interest rates this year has been a big headwind for technology stocks given their high-duration profile (i.e., secular cash flow producers). That said, our work on the technology sector finds the stocks attractive given their lower valuations and earnings advantage. We think they represent a unique opportunity to buy high profit margin businesses at a big discount.

The stocks are valued much lower than the Nifty Fifty cohort of yesteryear, that succumbed to very high valuations going into the 1970s inflation regime. Moreover, today’s tech stocks are much less risky than their 1990s predecessors, as their fundamentals are a lot more stable.

We’ve also found technology stocks among the leading sectors coming out of market drawdowns. We expect them to outperform after interest rates peak.

Transportation Stocks (Long, IYT)

In our prior work on the consumer, we highlighted the constructive case for transportation stocks. They stand to benefit in a disinflationary regime that see oil prices subside quickly, boosting discretionary spending from the consumer toward travel (i.e., airlines) and goods (i.e., trucking logistics, rails).

Falling oil prices would also bring down fuel costs for these oil-intensive industries, boosting profit margins. In fact, transportation stocks were the best-performing sector across various disinflation episodes we studied going back to the 1940s.

Bitcoin (Long, BTC)

Our work on Bitcoin reveals that it’s a levered bet on the S&P 500 (positively-correlated) and the dollar (negatively-correlated). As we mentioned earlier, we have a constructive view on stocks based on our expectations that interest rates will fall and earnings growth will be better than expected. We have a negative bias on the dollar related to our view that the Fed’s runway for more aggressiveness is now limited after reaching a neutral stance. As a result, we believe there’s upside in the cryptocurrency.

However, our research also points out that Bitcoin tends to underperform stocks following downturns, as investors look for confirmation that optimism in equities comes back. For instance, we’ve found that the best returns to Bitcoin came during low equity volatility regimes. While we seek out to anticipate the rebounds in stocks, as a big part of the returns accrue quickly, that’s less the case with Bitcoin. With the latter, we need the green light before increasing the bet.

Moreover, the dollar could continue rising if global uncertainty worsens from here, as that’s a critical risk factor in driving the greenback’s strength.

Crude Oil (Short, OIL)

Our disinflation thesis was highlighted in research published in June, with falling oil prices a large component of it. Back then we took note of the 130% rise in gasoline prices since the bottom in inflation seen in May 2020. We saw that as overextended, limiting the runway for further price gains. That’s because gasoline prices can be self-limiting, as rising costs can slow down demand.

We also highlighted that the jump in gasoline prices associated with the Ukraine/Russia war far exceeded prior run-ups that also featured a major oil supplier as a combatant. As a result, further upside in prices would need to see another oil shock in the making. Since then, weakening demand and a fading supply shock have pulled oil prices down, acting as a drag on inflation. The precipitous decline in oil prices of late has reduced the downside risk of oil for shorts, and therefore the risk-reward is now less favorable.

Treasury Bonds (Long, TLT)

Our sizeable allocation to stocks exposes us to weaker-than-expected economic growth that could undermine earnings. That view was detailed in prior research that highlighted the case for Treasury bonds as cheap portfolio insurance, following their large decline this year resulting from the Fed’s tightness. As inflation recedes from here, we expect interest rates to decline as the bond market anticipates the Fed changing its stance, representing a tailwind for bonds.

Our work on quantitative tightening didn’t find a clear relationship between a contracting balance sheet and bond returns, hence we don’t see it as a big threat.

Our allocation to bonds is based on its hedging profile, and it’s not a tactical bet. Therefore, its weight in our portfolio is a function of the current level of the yield curve. When the yield curve is inverted as much as today, bonds tend to do well in the next year.

Cash (BIL)

Our initial cash allocation is set at 10%. That’s high relative to our long-run target of 5%. A higher initial allocation at launch gives us flexibility to fund upcoming opportunities, or add to current ones, without having to reduce existing positions.

Conclusion: Next Steps

Our model portfolio reflects our highest conviction ideas backed up by our ongoing research, and it’s shown in the table below. Our biggest bet today is on stocks, that we see as leading beneficiaries as inflation recedes from here, interest rates fall and earnings prove resilient. We don’t foresee a deep recession at this point, but are hedging our equity exposure through Treasury bonds.

We’ll be updating our portfolio as our research evolves, adding new ideas and removing those that lose attractiveness. We’ll be sure to communicate those changes and provide our reasoning. We’ll also rebalance the portfolio over time to re-calibrate our exposures.

We’ve put a lot of effort into constructing our model portfolio and are very excited to launch it. We believe it’ll be a big value-add to our members in helping to inform their decision-making.

Here is today’s YouTube video: For the 5th time in Stock Market History… the SP500 Will See a MASSIVE Recovery.

 

Footnotes:

[1] The idea is to take into account what the market’s expectations for risk going forward are, while also weighing in volatility’s mean-revering property. We believe that’s a sensible approach because high implied volatilities can unfairly penalize estimates of risk-reward, particularly for high-conviction ideas.

[2] The optimizer uses an algorithm that iteratively varies the weights of the various asset classes in the portfolio to find the combination offering the highest expected return per unit of risk. The optimizer also needs a correlation matrix in order to have a sense for how each asset co-moves with the other, assessing diversification benefits. We use a weighted-average of both short-and long-term correlations.